We have developed a model for polyarticular arthritis in BALB/C mice which is initiated by immunization with "arthritogenic" proteoglycans isolated from human osteophytes, human chondrosarcoma and fetal pig articular cartilage. This model has appealing features in that the degenerative disease process can be transmitted by the passage of lymphocytes from arthritic animals to immunosuppressed, non-immunized naive animals. An understanding of the antigenic determinant(s) which are involved in the initiation of polyarthritis would provide valuable insight into biochemical structures which might interact with the immune system to produce a continuing degenerative process. Potential candidates for the arthritogenic fragments of proteoglycans will be selected based on multiple criteria. Prime candidates should exhibit reactivity with monoclonal antibodies against mouse cartilage proteoglycan and with immune sera obtained from inoculated mice both prior and subsequent to overt inflammation and degeneration but not with immune sera from animals injected with non-arthritogenic proteoglycans. We have also shown that both T-lymphocytes and antibodies against proteoglycans are required for the propagation of the disease. T-cell clones which react positively with proteoglycan fragments which are arthritogenic or which differentially stimulate lymphocytes from arthritic animals but not those from normal animals will be isolated. Antibodies and T-cell clones will be used to generate progressive polyarthritis in immuno-suppressed syngeneic mice. Fragments of proteoglycans containing arthritogenic sites will be examined further for the specific structure which may be responsible for the initiation of arthritis. These experiments will include a determination of the amino acid sequence and post-translation modifications to the core protein. Ultimately, synthetic peptides will be used to identify the proteoglycan structure that is critical for interaction with T-cell receptors.